Menu Sign In Contact FAQ
Banner
Welcome to our forums

Two alternators, two batteries?

10 Posts

It is a question I have long been wondering over: if a plane has two alternators, either on one engine or on two, can it make do with only a single battery?

I cannot see two alternators feeding/charging a single battery, how can they ever be made to supply EXACTLY the same voltage? And if they don't, there'll be only a solid current from one alternator to the other, not what one really wants.

OTOH the perfect installation would have two fully separate circuits, but that would require two batteries too and that would be more weight than really required.

How is this done, either on singles with a secondary alternator on the one engine (as I understand is required for IFR certification) or on a twin with an alternator on each engine?

I just hope nobody comes to describe a scheme with four alternators (two on each engine) and three batteries...?

EBZH Kiewit, Belgium

The PA-46 has two engine driven alternators. They each have voltage regulators and feed into a tie bus which then feeds the various systems. It provides redundancy in that you can run much more load than with one and if one fails you can run a still significant load on a single alternator. Obviously if they both fail then two batteries would help but that is not the system I have.

EGTK Oxford

...as I understand is required for IFR certification...

No, that's not a requirement for IFR certification! I instruct IFR on quite a few different IFR certified singles none of which has two alternators. If a single alternator fails in IMC, that's no big deal anyway in the year 2012. Even complete electrical failure is nothing to worry about as long as you have an independent communication device (handheld radio or cellphone) and a GPS receiver - the one in your smartphone is sufficient to bring you home.

The piece of kit that connects two alternators to one battery is called "paralleling unit". It matches the output voltage of the two alternators by modulating theIr field current. Unfortunately, this is the weak spot of the system because if this single unit fails, you lose both (otherwise perfectly working!) generators together. Been there, done that - in a piston twin. The handheld radio had to save yet another day.

EDDS - Stuttgart

...as I understand is required for IFR certification...

No, that's not a requirement for IFR certification!

It is required by the FAA for FIKI (flight into known icing) certification whereas it's not required by EASA. There are a number of airplanes with FIKI in Europe that would not get FIKI approval in the US.

I cannot see two alternators feeding/charging a single battery, how can they ever be made to supply EXACTLY the same voltage?

That's not terribly hard to do, there are voltage regulators and diodes. Also we're talking DC here.

I instruct IFR on quite a few different IFR certified singles none of which has two alternators.

The more electric your airplane is, the more likely you are going to have two alternators. In a single alternator airplane, the battery is considered to be the backup system of the alternator. You really don't want to lose your AI if your alternator packs up.

The Cirrus is a great example of a modern all electric system with a well designed redundancy. This accident report is a very interesting read about such systems and the need to understand them as pilot.

Making some enquiries, it appears there are several ways of doing this system.

The Cirrus has two alternators and two voltage regulators, one set to 28.5V and the other set to 27.5V.

Another system is here. This appears to have a manual switchover to the 2nd alternator. Other approaches are among this lot.

The problem is that the 3-phase rectified output has a lot of ripple - of the order of 50% - so a battery is needed on the alternator's output to get a clean DC.

This is why the 2-alternator 1-battery setups have the sometimes complicated switching, instead of the obvious way which is just having two 28.6V alternators feeding 2 diodes to deliver a single 28V output.

The GAMI alternator appears to modulate its field current deeply enough and fast enough to not need a battery (they have been demonstrating it as exhibitions thus) but I don't think any production "main alternator" system does that.

You really don't want to lose your AI if your alternator packs up.

There are strict approval requirements for the AI, for a single-alternator plane. An electric AI is acceptable if it has a backup battery but the battery has to be quite large - IIRC more than 30 mins' worth. This is why a vacuum instrument is popular - it's an easy way out, despite having to change the filters regularly.

Administrator
Shoreham EGKA, United Kingdom

Had that double failure also in a twin. Was in FL120 in severe CAVOK and tried to manage the flight with a portable radio and handheld gps. It worked out, but:

  • the portable radio does quite nothing with the portable antenna, distance to the controllers antenna was to far. External Antenna would have been better.
  • autopilot fails instantly, at night you would have no lighting at all, communication to the passengers is ZERO - and of course now they have their questions!

All this in cavok with a 14 year old guy on the left seat who manages to hold the plane on course and at altitude is a piece of cake.

Alone at night with "full house", you need a good capacity management!

(sorry for going slightly off topic)

One six right!
EDVE

Another system is here. This appears to have a manual switchover to the 2nd alternator.

Sorry to disagree: as I read that one, the second alternator uses a special regulator, with a Hall effect sensor measuring the current.

The link to aeroelectric is a gold mine, thanks for sharing!

EBZH Kiewit, Belgium

Thinking and rethinking, I have come to realise that alternators form their output by adding up the output of three windings through 3 diodes. So there can never be a current from one alternator into the other. Still no way to make sure they will share the load 50/50, except by measuring the current with a shunt or a Hall sensor

EBZH Kiewit, Belgium

Sorry to disagree: as I read that one, the second alternator uses a special regulator, with a Hall effect sensor measuring the current.

There is certainly a switch provided to enable the 2nd alternator.

I can't say one way or the other whether the 2nd alternator is meant to be enabled all the time.

One could not have it energised continuously and drive the field current purely from sensing the output current unless the output current target was variable according to some rule. If one was driving the field to achieve e.g. a constant 10A then as soon as the aircraft total demand fell below 10A, the bus voltage would be driven up uncontrollably and you would end up pumping up to 10A into the battery, which would not do it much good after it became fully charged.

So I don't know how it works... there has to be a voltage target as in a normal alternator, but in purely electrical terms you can't have 2 or more voltage sources in parallel if you want load sharing. It can certainly be done with current measurement (I guess one way would be to drive the field for a constant current but impose a voltage limit) and maybe the voltage regulator with the current sense does something like that.

I have come to realise that alternators form their output by adding up the output of three windings through 3 diodes. So there can never be a current from one alternator into the other

Yes; absolutely right.

The problem is that the crude voltage regulators in common use don't sense the output voltage before the rectifier. If they did, parallelling alternators would be trivial (just make sure the output voltages are accurately controlled so the output impedance of the alternator ensures load sharing all by itself, and wire the outputs up all together). They sense the voltage after the rectifier, which is mega smoothed with the assistance of the battery acting as a huge capacitor.

except by measuring the current with a shunt or a Hall sensor

and have the voltage regulators intercommunicate so they load share by equalising the currents, which would be really smart, but AFAICT that is not happening...

The link to aeroelectric is a gold mine, thanks for sharing!

I have a massive collection of avionics manuals Been collecting them off the internet (and other places) for years. Can't post the URL openly though because the server is a very slow one. You know where to find me

Had that double failure also in a twin.

A pilot I know had a total loss of electrics in a Twin Com, where a wire came off the back of the Master switch.

the portable radio does quite nothing with the portable antenna, distance to the controllers antenna was to far. External Antenna would have been better.

I got that done when my plane was new. Got the coax to the COM2 antenna looped via a couple of BNC sockets, about 2" apart, which are normally interconnected with a short cable loop. The Icom radio plugs into one of these. The range, AFAICT, is almost the same as the normal radio (KX1x5A). Actually one of the two sockets is an SMA, not a BNC, to ensure that the Icom radio can be plugged only into the right one

Administrator
Shoreham EGKA, United Kingdom

have the voltage regulators intercommunicate so they load share by equalising the currents, which would be really smart, but AFAICT that is not happening...

Had just caught myself scribbling two microcontrollers, each steering an alternator field through a switching MOSFET and sensing alternator current through the shunt you want anyway for an ammeter... and talking to one another by I2C or whatever. Can't be really hard to do. Oh well, one more project on the "to-do" list, with a VERY low priority, as I can't see myself flying multi-engine anywhere soon. Although, perhaps, who knows...:

EBZH Kiewit, Belgium
10 Posts
Sign in to add your message

Back to Top